krutrim-admin
commited on
Updated Indic evals
Browse files
README.md
CHANGED
@@ -94,10 +94,10 @@ We use the LM Evaluation Harness to evaluate our model on the En benchmarks task
|
|
94 |
|--------------------------------------------|------------|--------------|----------------|--------------|--------------|----------------|--------|
|
95 |
| IndicSentiment (0-shot) | Accuracy | 0.65 | 0.70 | 0.95 | 0.96 |0.99 | 0.98 |
|
96 |
| IndicCOPA (0-shot) | Accuracy | 0.51 | 0.58 | 0.80 | 0.83 | 0.88 | 0.91 |
|
97 |
-
| IndicXParaphrase (0-shot) | Accuracy | 0.67 | 0.74 | 0.88 | 0.87 | 0.89 |
|
98 |
-
| IndicXNLI (0-shot) | Accuracy | 0.47 | 0.54 | 0.55 |
|
99 |
| IndicQA (0-shot) | Bert Score | 0.90 | 0.90 | 0.91 | TBD | TBD | TBD |
|
100 |
-
| CrossSumIN (1-shot) | chrF++ | 0.04 | 0.17 | 0.21 | 0.26 | 0.24 |
|
101 |
| FloresIN Translation xx-en (1-shot) | chrF++ | 0.54 | 0.50 | 0.58 | 0.60 | 0.62 | 0.63 |
|
102 |
| FloresIN Translation en-xx (1-shot) | chrF++ | 0.41 | 0.34 | 0.48 | 0.46 | 0.47 | 0.48 |
|
103 |
| IN22 Translation xx-en (0-shot) | chrF++ | 0.50 | 0.48 | 0.57 | 0.58 | 0.55 | 0.55 |
|
|
|
94 |
|--------------------------------------------|------------|--------------|----------------|--------------|--------------|----------------|--------|
|
95 |
| IndicSentiment (0-shot) | Accuracy | 0.65 | 0.70 | 0.95 | 0.96 |0.99 | 0.98 |
|
96 |
| IndicCOPA (0-shot) | Accuracy | 0.51 | 0.58 | 0.80 | 0.83 | 0.88 | 0.91 |
|
97 |
+
| IndicXParaphrase (0-shot) | Accuracy | 0.67 | 0.74 | 0.88 | 0.87 | 0.89 | 0.91 |
|
98 |
+
| IndicXNLI (0-shot) | Accuracy | 0.47 | 0.54 | 0.55 | 0.61 | TBD | 0.67 |
|
99 |
| IndicQA (0-shot) | Bert Score | 0.90 | 0.90 | 0.91 | TBD | TBD | TBD |
|
100 |
+
| CrossSumIN (1-shot) | chrF++ | 0.04 | 0.17 | 0.21 | 0.26 | 0.24 | 0.24 |
|
101 |
| FloresIN Translation xx-en (1-shot) | chrF++ | 0.54 | 0.50 | 0.58 | 0.60 | 0.62 | 0.63 |
|
102 |
| FloresIN Translation en-xx (1-shot) | chrF++ | 0.41 | 0.34 | 0.48 | 0.46 | 0.47 | 0.48 |
|
103 |
| IN22 Translation xx-en (0-shot) | chrF++ | 0.50 | 0.48 | 0.57 | 0.58 | 0.55 | 0.55 |
|